The Institute for Electrical and Electronics Engineers (IEEE), in resolution IEEE 802.11, also referred as “802.11”, has defined a plurality of specifications which are related to wireless networking. Among them are specifications for “closed loop” feedback mechanisms by which a receiving mobile terminal may feed back information to a transmitting mobile terminal to assist the transmitting mobile terminal in adapting signals which are sent to the receiving mobile terminal.
Smart antenna systems combine multiple antenna elements with a signal processing capability to optimize the pattern of transmitted signal radiation and/or reception in response to the communications medium environment. The process of optimizing the pattern of radiation is sometimes referred to as “beamforming,” which may utilize linear array mathematical operations to increase the average signal to noise ratio (SNR) by focusing energy in desired directions. In conventional smart antenna systems, only the transmitter or the receiver may be equipped with more than one antenna, and may typically be located in the base transceiver station (BTS) where the cost and space associated with smart antenna systems have been perceived as more easily affordable than on mobile terminals such as cellular telephones. Such systems are also known as multiple input single output (MISO) when a multiple antenna transmitter is transmitting signals to a single antenna receiver, or single input multiple output (SIMO) when a multiple antenna receiver is receiving signals that have been transmitted from a single antenna transmitter. With advances in digital signal processing (DSP) integrated circuits (ICs) in recent years, multiple antenna multiple output (MIMO) systems have emerged in which mobile terminals incorporate smart antenna systems comprising multiple transmit antenna and multiple receive antenna. One area of early adoption of MIMO systems has been in the field of wireless networking, particularly as applied to wireless local area networks (WLANs) where transmitting mobile terminals communicate with receiving mobile terminals. IEEE resolution 802.11 comprises specifications for communications between mobile terminals in WLAN systems.
Signal fading is a significant problem in wireless communications systems, often leading to temporary loss of communications at mobile terminals. One of the most pervasive forms of fading is known as multipath fading, in which dispersion of transmitted signals due to incident reflections from buildings and other obstacles, results in multiple versions of the transmitted signals arriving at a receiving mobile terminal. The multiple versions of the transmitted signal may interfere with each other and may result in a reduced signal level detected at the receiving mobile terminal. When versions of the transmitted signal are 180° out of phase they may cancel each other such that a signal level of 0 is detected. Locations where this occurs may correspond to “dead zones” in which communication to the wireless terminal is temporarily lost. This type of fading is also known as “Rayleigh” or “flat” fading.
A transmitting mobile terminal may transmit data signals in which data is arranged as “symbols”. The transmission of symbols may be constrained such that after a symbol is transmitted, a minimum period of time, Ts, must transpire before another symbol may be transmitted. After transmission of a symbol from a transmitting mobile terminal, some period of dispersion time, Td, may transpire which may be the time over which the receiving mobile terminal is able to receive the symbol, including multipath reflections. The time Td may not need to account for the arrival of all multipath reflections because interference from later arriving reflected signals may be negligible. If the period Ts is less than Td there is a possibility that the receiving mobile terminal will start receiving a second symbol from the transmitting mobile terminal while it is still receiving the first symbol. This may result in inter-symbol interference (ISI), producing distortion in received signals, and possibility resulting in a loss of information. The quantity 1/Td is also referred to as the “coherence bandwidth” which may indicate the maximum rate at which symbols, and correspondingly information, may be transmitted via a given communications medium. One method to compensate for ISI in signals may entail utilizing DSP algorithms which perform adaptive equalization.
Another important type of fading is related to motion. When a transmitting mobile terminal, or a receiving mobile terminal is in motion, the Doppler phenomenon may affect the frequency of the received signal. The frequency of the received signal may be changed by an amount which is a function of the velocity at which a mobile terminal is moving. Because of the Doppler effect, ISI may result when a mobile terminal is in motion, particularly when the mobile terminal is moving at a high velocity. Intuitively, if a receiving mobile terminal is in motion and nearing a transmitting mobile terminal, the distance between the two mobile terminals will change as a function of time. As the distance is reduced, the propagation delay time, Tp, which is the time between when a transmitter first transmits a signal and when it first arrives at a receiver, is also reduced. As the mobile terminals become closer it is also possible that Td may be increased if, for example, the transmitting mobile terminal does not reduce the radiated power of transmitted signals. If Tp becomes less than Td, there may be ISI due to the Doppler effect. This case, which illustrates why data rates may be reduced for mobile terminals that are in motion, is referred to as “fast fading”. Because fast fading may distort signals at some frequencies while not distorting signals at other frequencies, fast fading may also be referred to as “frequency selective” fading.
Smart antenna systems may transmit multiple versions of a signal in what is known as “spatial diversity”. A key concept in spatial diversity is that the propagation of multiple versions of a signal, or “spatial stream”, from different antenna may significantly reduce the probability of flat fading at the receiving mobile terminal since not all of the transmitted signals would have the same dead zone.
Current transmission schemes in MIMO systems typically fall into two categories: data rate maximization, and diversity maximization. Data rate maximization focuses on increasing the aggregate data transfer rate between a transmitting mobile terminal and a receiving mobile terminal by transmitting different spatial streams from different antenna. One method for increasing the data rate from a transmitting mobile terminal would be to decompose a high bit rate data stream into a plurality of lower bit rate data streams such that the aggregate bit rates among the plurality of lower bit rate data streams is equal to that of the high bit rate data stream. Next, each of the lower bit rate data streams may be mapped to at least one of the transmitting antenna for transmission. In addition, each signal comprising one of the lower bit rate data streams is multiplicatively scaled by a weighting factor prior to transmission. The plurality of multiplicative scale factors applied to the plurality of signals comprising the lower bit rate data streams may be utilized to form the transmitted “beam” in the beamforming technique. An example of a data rate maximization scheme is orthogonal frequency division multiplexing (OFDM), in which each of the plurality of signals is modulated by a different frequency carrier signal prior to mapping and multiplicative scaling. OFDM transmission may be resistant to multipath fading in that a portion, but most likely not all, of the data transmitted may be lost at any instant in time due to multipath fading.
Diversity maximization focuses on increasing the probability that a signal transmitted by a transmitting mobile terminal will be received at a receiving mobile terminal, and on increasing the SNR of received signals. In diversity maximization, multiple versions of the same signal may be transmitted by a plurality of antenna. The case in which a transmitting mobile terminal is transmitting the same signal via all of its transmitting antenna may be the pure spatial diversity case in which the aggregate data transfer rate may be equal to that of a single antenna mobile terminal. There is a plurality of hybrid adaptations of the data rate and spatial diversity maximization schemes which achieve varying data rates and spatial diversities.
MIMO systems employing beamforming may enable the simultaneous transmission of multiple signals occupying a shared frequency band, similar to what may be achieved in code division multiple access (CDMA) systems. For example, the multiplicative scaling of signals prior to transmission, and a similar multiplicative scaling of signals after reception, may enable a specific antenna at a receiving mobile terminal to receive a signal which had been transmitted by a specific antenna at the transmitting mobile terminal to the exclusion of signals which had been transmitted from other antenna. However, MIMO systems may not require the frequency spreading techniques used in CDMA transmission systems. Thus, MIMO systems may make more efficient utilization of frequency spectrum.
One of the challenges in beamforming is that the multiplicative scale factors which are applied to transmitted and received signals may be dependent upon the characteristics of the communications medium between the transmitting mobile terminal and the receiving mobile terminal. A communications medium, such as a radio frequency (RF) channel between a transmitting mobile terminal and a receiving mobile terminal, may be represented by a transfer system function, H. The relationship between a time varying transmitted signal, x(t), a time varying received signal, y(t), and the systems function may be represented as shown in equation [1]:y(t)=H×x(t)+n(t), where  equation[1]n(t) represents noise which may be introduced as the signal travels through the communications medium and the receiver itself. In MIMO systems, the elements in equation[1] may be represented as vectors and matrices. If a transmitting mobile terminal comprises M transmitting antenna, and a receiving mobile terminal comprises N receiving antenna, then y(t) may be represented by a vector of dimensions Nx1, x(t) may be represented by a vector of dimensions Mx1, n(t) by a vector of dimensions Nx1, and H may be represented by a matrix of dimensions N×M. In the case of fast fading, the transfer function, H, may itself become time varying and may thus also become a function of time, H(t). Therefore, individual coefficients, hij(t), in the transfer function H(t) may become time varying in nature.
In MIMO systems which communicate according to specifications in IEEE resolution 802.11, the receiving mobile terminal may compute H(t) each time a frame of information is received from a transmitting mobile terminal based upon the contents of a preamble field in each frame. The computations which are performed at the receiving mobile terminal may constitute an estimate of the “true” values of H(t) and may be known as “channel estimates”. For a frequency selective channel there may be a set of H(t) coefficients for each tone that is transmitted via the RF channel. To the extent that H(t), which may be referred to as the “channel estimate matrix”, changes with time and to the extent that the transmitting mobile terminal fails to adapt to those changes, information loss between the transmitting mobile terminal and the receiving mobile terminal may result.
Higher layer communications protocols, such as the transmission control protocol (TCP) may attempt to adapt to detected information losses, but such adaptations may be less than optimal and may result in slower information transfer rates. In the case of fast fading, the problem may actually reside at lower protocol layers, such as the physical (PHY) layer, and the media access control (MAC) layer. These protocol layers may be specified under IEEE 802.11 for WLAN systems. The method by which adaptations may be made at the PHY and MAC layers, however, may comprise a mechanism by which a receiving mobile terminal may provide feedback information to a transmitting mobile terminal based upon channel estimates which are computed at the receiving mobile terminal.
Existing closed loop receiver to transmitter mechanisms, also referred as “RX to TX feedback mechanisms”, that exist under IEEE 802.11 include acknowledgement (ACK) frames, and transmit power control (TPC) requests and reports. The TPC mechanisms may allow a receiving mobile terminal to communicate information to a transmitting mobile terminal about the transmit power level that should be used, and the link margin at the receiving mobile terminal. The link margin may represent the amount of signal power that is being received, which is in excess of a minimum power required by the receiving mobile terminal to decode message information, or frames, that it receives.
A plurality of proposals is emerging for new feedback mechanisms as candidates for incorporation in IEEE resolution 802.11. Among the proposals for new feedback mechanisms are proposals from TGn (task group N) sync, which is a multi-industry group that is working to define proposals for next generation wireless networks which are to be submitted for inclusion in IEEE 802.11, and Qualcomm. The proposals may be based upon what may be referred as a “sounding frame”. The sounding frame method may comprise the transmitting of a plurality of long training sequences (LTSs) that match the number of transmitting antenna at the receiving mobile terminal. The sounding frame method may not utilize beamforming or cyclic delay diversity (CDD). In the sounding frame method, each antenna may transmit independent information.
The receiving mobile terminal may estimate a complete reverse channel estimate matrix, Hup, for the channel defined in an uplink direction from the receiving mobile terminal to the transmitting mobile terminal. This may require calibration with the transmitting mobile terminal where the transmitting mobile terminal determines the forward channel estimate matrix, Hdown, for the channel defined in a downlink direction from the transmitting mobile terminal to the receiving mobile terminal. To compensate for possible differences between Hup and Hdown the receiving mobile terminal may be required to receive Hdown from the transmitting mobile terminal, and to report Hup−Hdown as feedback information. The TGn sync proposal may not currently define a calibration response. A channel estimate matrix may utilize 24 or more bits for each channel and for each tone, comprising 12 or more bits in an in-phase (I) component and 12 or more bits in a quadrature (Q) component.
According to the principle of channel reciprocity, the characteristics of the RF channel in the direction from the transmitting mobile terminal to the receiving mobile terminal may be the same as the characteristics of the RF channel in the direction from the receiving mobile terminal to the transmitting mobile terminal Hup=Hdown. In actual practice, however, there may be differences in the electronic circuitry between the respective transmitting mobile terminal and receiving mobile terminal such that, in some cases, there may not be channel reciprocity. This may require that a calibration process be performed in which Hup and Hdown are compared to reconcile differences between the channel estimate matrices. However, there may be limitations inherent in some calibration processes. For example, some proposals for new IEEE 802.11 feedback mechanisms may be limited to performing “diagonal calibrations”. These methods may not be able to account for conditions in which there are differences in non-diagonal coefficients between Hup and Hdown. These non-diagonal coefficient differences may be the result of complicated antenna couplings at the respective transmitting mobile terminal and/or receiving mobile terminal. Accordingly, it may be very difficult for a calibration process to correct for these couplings. The ability of a calibration technique to accurately characterize the RF channel at any instant in time may be dependent upon a plurality of dynamic factors such as, for example, temperature variations. Another limitation of calibration procedures is that it is not known for how long a calibration renders an accurate characterization of the RF channel. Thus, the required frequency at which the calibration technique must be performed may not be known.
Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art, through comparison of such systems with some aspects of the present invention as set forth in the remainder of the present application with reference to the drawings.